Dynamic Marketing Mix Modeling
Our dynamic MMM service focuses on aggregated groups of consumers ranging from total market through to store and household panel with a focus on:
Short-term media effectiveness
In the modern digital economy, offline and online marketing activity are inherently linked in the product purchase decision, where paid, owned and earned media combine to drive incremental volumes. One of the main challenges facing clients today is how best to measure media effectiveness in this environment. More specifically:
- What is the ROI on my off and online marketing investments?
- How do I allocate my off and online marketing budget more efficiently?
- What is the impact of competitor actions and vice versa?
- How do I quantify the category growth effects of my media strategy?
- What is the combined impact of paid, owned and earned media?
- How should I design my off and online media strategy?
- What is the value of social media?
- How do I predict future consumer demand for my products?
To answer these questions, we apply dynamic state-space MMM approaches combined with large-scale Hierarchical Bayesian techniques. This allows us to cleanly separate short and long-term demand behavior from SKU to market level.
Long-term media effectiveness
Models that focus solely on incremental volumes ignore the long-run view: that is, the potential long-term brand-building properties of successful media campaigns. To address this issue typically requires additional information on the role of brand perceptions in driving long-term behavior - ranging from unaided awareness, consideration and NPS scores, through to brand likeability and experiences. This helps answer questions such as:
- How do customer perception metrics, PR, earned media and long-term base demand drive each other to build brand?
- What is the long-term (brand-building) impact of my paid media campaigns?
- How do I build brand equity?
To answer these questions, we focus on the long-term demand patterns isolated during dynamic MMM analysis – creating cointegrated VAR (network) models of long-term base sales, consumer brand perceptions, paid and earned media. The result uncovers the true drivers of long-term brand building.
Pricing plays a central role in consumer demand analysis across all sectors of the economy, from Consumer Packaged Goods, airlines and entertainment events through to financial services. A key challenge is how best to set prices for optimal sales and brand performance. More specifically:
- What is my optimal short-term discount level?
- How can I set prices to minimize brand equity erosion?
- How can I set pricing dynamically over days and weeks to optimize sales and revenue targets?
- How do I best estimate brand, UPC and SKU level price response for my products and brands?
- What is the optimal interest rate strategy for new and existing customers?
These questions are addressed as part of our dynamic MMM service, with depth of advice depending on the granularity of analysis. At high levels of aggregation, pricing response is more contaminated by distributional effects and price is more of a general control factor. At lower levels – such as sales channel and product – (dynamic) price elasticities are more actionable for optimal price-setting.
At Marketscience, we have combined detailed media and pricing analytics into Marketscience Studio: a fully transparent demand measurement platform, combining the economics of consumer behavior, network analysis and long-term brand-building.